What features does Image Moderation Enhanced Edition offer, and how is it billed?-AI Guardrails(AI Guardrails)-阿里云帮助中心

Image Moderation Enhanced Edition detects risky image content across multiple scenarios and supports pay-as-you-go and resource plan billing.

1. Introduction to Image Moderation Enhanced Edition

Features

The Image Moderation 2.0 API detects image content that violates online regulations, disrupts platform governance, or degrades user experience. It supports 90+ content risk tags and 100+ risk control items. Based on the risk tags and confidence scores returned by the API, you can define custom review or governance actions tailored to your industry standards or platform content policies.

Version comparison

Compared with Image Moderation 1.0, Image Moderation Enhanced Edition offers broader risk coverage, more granular risk labels, and more flexible console configuration options.

Comparison item	Image ModerationEnhanced Edition	Image Moderation 1.0
Default QPS	Non-large-model version: 100 Large-model version: 50	50
Risk detection scope	LLM moderation: Image Moderation for Large and Small Model Integration: Combines large-model and expert-model capabilities to detect pornography, sexy content, politically sensitive content, terrorist content, prohibited content, religious content, advertising traffic diversion, inappropriate content, and other risks. Image Moderation for Large and Small Model Integration_For regions outside the Chinese mainland: For international use cases, combines large-model and expert-model capabilities to detect pornography, sexy content, politically sensitive content, terrorist content, prohibited content, religious content, flags, advertising traffic diversion, inappropriate content, abuse, and other violations. Image Moderation Service Based on LLMs: Uses a LLM to detect pornography, politically sensitive content, terrorist content, prohibited content, inappropriate content, abuse, advertising, and other risks. Large Model-Powered Ad Traffic Detection: Uses large-model capabilities to accurately identify adversarial and AI-generated advertising traffic diversion content. General-purpose scenarios: BaselineCheck: Detects pornography (including text-image pornography), sexy content, politically sensitive content (including text-image politically sensitive content), terrorist content (including text-image terrorist content), and prohibited content (including text-image prohibited content). baselineCheck_pro: Returns more granular labels than Baseline Check. Common Baseline Moderation_For regions outside the Chinese mainland: Detects pornography (including text-image pornography), sexy content, politically sensitive content (including text-image politically sensitive content), terrorist content (including text-image terrorist content), prohibited content (including text-image prohibited content), flags, inappropriate guidance, abuse, and specific scenarios. TonalityImprove: Detects advertising traffic diversion, advertising law violations, inappropriate guidance, abuse, special logos, specific objects, and image quality issues. Content Governance Moderation_For regions outside the Chinese mainland: Detects advertising traffic diversion, inappropriate guidance, special logos, and behavioral content. AI-Generated Content (AIGC) scenarios: AIGC Image Risk Check: Detects pornography, vulgar content, sexy content, politically sensitive content, terrorist content, prohibited content, disturbing content, and inappropriate content. AIGC Image Risk Detection (Outside Chinese Mainland): Detects pornography, vulgar content, sexy content, politically sensitive content, terrorist content, prohibited content, disturbing content, and inappropriate content. AIGC Violation Detection: Detects trademarks, special logos, people, and other potential infringement elements in images. AIGC image detection: Determines whether an image is likely generated by AIGC. AI-Generated Image Detection_Implicit Identifier: Adds implicit marker detection to AIGC image identification. AI Image Detection (Video Screenshots): For video use cases, detects whether video frames are likely AI-generated or synthetic. AIGC Detection_Professional Edition: Determines whether an image is likely generated by AIGC and whether it shows signs of synthesis (including face synthesis). AI-Generated Image Detection_Ultimate: Determines whether an image is likely generated by AIGC, whether it shows signs of synthesis (including face synthesis), and whether it contains AI-based local edits. Business scenarios: Profile Photo Check: Detects pornography, sexy content, politically sensitive content, terrorist content, prohibited content, religious content, advertising traffic diversion, special logos, behaviors, specific objects, inappropriate content, abuse, and image quality issues. Post Comment Image Moderation: Detects pornography, sexy content, politically sensitive content, terrorist content, prohibited content, religious content, advertising traffic diversion, special logos, behaviors, specific objects, and inappropriate content. Advertising Check: Detects advertising law violations, prohibited content, pornography, sexy content, politically sensitive content, terrorist content, religious content, behaviors, inappropriate content, advertising traffic diversion, and specific objects. Live Stream Check: Detects pornography, sexy content, politically sensitive content, terrorist content, prohibited content, religious content, inappropriate content, advertising traffic diversion, behaviors, specific objects, and image quality issues. Special scenarios: Moderate Images for Malicious Content: Detects whether an image hides video segments or players.	Pornography, politically sensitive and terrorist content, text-image violations, inappropriate scenes, special logos, QR codes
Risk detection labels	100+ Note For a full list of labels, see the Risk label interpretation table.	40+ Note For a full list of labels, see label.
Console	Configure detection items Set up custom image libraries Set up custom dictionaries Copy services (for multiple business needs) Query results for up to 30 days Note For more information, see the Console operation guide.	Configure detection items Set up custom image libraries Query results for up to 7 days
Supported image formats	PNG, JPG, JPEG, BMP, WEBP, TIFF, SVG, GIF, ICO, HEIC	PNG, JPG, JPEG, BMP, GIF, WEBP
Billing	Billed per service. Each service detects multiple risks simultaneously. Cost = Number of images × Number of services Billed separately for the following services: BaselineCheck baselineCheck_pro Common Baseline Moderation_For regions outside the Chinese mainland TonalityImprove Content Governance Moderation_For regions outside the Chinese mainland Image Moderation for Large and Small Model Integration Image Moderation for Large and Small Model Integration_For regions outside the Chinese mainland Image Moderation Service Based on LLMs LLMLarge Model-Powered Ad Traffic Detection AIGC Image Risk Check AIGC Image Risk Detection (Outside Chinese Mainland) AIGC Violation Detection AIGC image detection AI-Generated Image Detection_Implicit Identifier AI Image Detection (Video Screenshots) AIGC Detection_Professional Edition AI-Generated Image Detection_Ultimate Profile Photo Check Post Comment Image Moderation Advertising Check Live Stream Check Moderate Images for Malicious Content	Billing is based on threat scenarios (scene). Billed per risk scenario (scene). Cost = Number of images × Number of risk scenarios Billed separately for the following risk scenarios: Image pornography detection Politically sensitive and terrorist content Text-image violations Inappropriate scenes Special logos QR codes

Services

Image Moderation Enhanced Edition supports the following services:

Scenario	Service	Detected content	Applicable scenarios
General-purpose scenarios	BaselineCheck (baselineCheck)	Detects high-risk content such as pornography, sexually explicit content, politically sensitive content, terrorist content, and prohibited content. Includes both visual content and embedded text. For a full list of detectable items, see the AI Guardrails Console.	Use this service to detect whether images contain noncompliant or unsuitable content. Apply it to all images exposed to the public internet.
	baselineCheck_pro (baselineCheck_pro)	Returns more granular labels than Baseline Check. For a full list of detectable items, see the AI Guardrails Console.	Detects prohibited or inappropriate content in images. Use this feature when you require fine-grained processing and have limited customization needs.
	Common Baseline Moderation_For regions outside the Chinese mainland (baselineCheck_cb)	For international use cases, detects pornography (including text-image pornography), sexually explicit content, politically sensitive content (including text-image politically sensitive content), terrorist content (including text-image terrorist content), prohibited content (including text-image prohibited content), flags, inappropriate content, abuse, and specific scenarios. Includes both visual content and embedded text in 18 languages: Chinese, English, French, German, Indonesian, Malay, Portuguese, Spanish, Thai, Vietnamese, Japanese, Arabic, Filipino, Hindi, Türkiye, Russian, Italian, and Dutch. For a full list of detectable items, see the AI Guardrails Console.	Use this service to detect whether images contain noncompliant or unsuitable content. Apply it to all images exposed to the public internet outside China.
	TonalityImprove (tonalityImprove)	Detects advertising traffic diversion, advertising law violations, inappropriate guidance, abuse, special logos, specific objects, and other governance-related content. Includes both visual content and embedded text. For a full list of detectable items, see the AI Guardrails Console.	Use this service to detect whether images contain content that disrupts platform order, harms content tone, or degrades user experience. Combine it with Baseline Check to meet governance requirements.
	Content Governance Moderation_For regions outside the Chinese mainland (tonalityImprove_cb)	For international use cases, detects advertising traffic diversion, inappropriate guidance, special logos, and behavioral content. Includes both visual content and embedded text. For a full list of detectable items, see the AI Guardrails Console.	Use this service to detect whether images contain content that disrupts platform order, harms content tone, or degrades user experience. Combine it with Baseline Check to meet governance requirements.
LLM moderation	Image Moderation for Large and Small Model Integration (postImageCheckByVL)	Combines large-model and expert-model capabilities to detect pornography, sexually explicit content, politically sensitive content, terrorist content, prohibited content, religious content, advertising traffic diversion, and inappropriate content. Returns detailed labels. For a full list of detectable items, see the AI Guardrails Console.	Use this service when high accuracy is required. For more information, see LLM-based image moderation enhanced service.
	Image Moderation for Large and Small Model Integration_For regions outside the Chinese mainland (postImageCheckByVL_cb)	For international use cases, combines large-model and expert-model capabilities to detect pornography, sexually explicit content, politically sensitive content, terrorist content, prohibited content, religious content, flags, advertising traffic diversion, inappropriate content, abuse, and other violations. Returns detailed labels. For a full list of detectable items, see the AI Guardrails Console.	Use this service when high accuracy is required. For more information, see LLM-based image moderation enhanced service.
	Image Moderation Service Based on LLMs (baselineCheckByVL)	Uses a LLM trained specifically for image moderation to detect pornography, politically sensitive content, terrorist content, prohibited content, inappropriate content, abuse, advertising, and other risks. Returns top-level labels. For a full list of detectable items, see the AI Guardrails Console.	Use this service to experience large-model moderation capabilities. For more information, see LLM-based image moderation enhanced service.
	LLMLarge Model-Powered Ad Traffic Detection (adCheckByVL)	Uses a custom-trained LLM to deeply understand image semantics and accurately identify highly concealed and adversarial advertising traffic diversion behaviors, including contact information camouflage, misleading redirects, QR code steganography, and AI-generated advertising traffic diversion.	Use this service in image moderation scenarios to accurately identify various adversarial and AI-generated advertising traffic diversion content. For more information, see LLM-based image moderation enhanced service.
AIGC scenarios	AIGC Image Risk Check (aigcCheck)	Optimized for AI-generated images. Detects pornography, vulgar content, sexually explicit content, politically sensitive content, terrorist content, prohibited content, disturbing content, and inappropriate content. For a full list of detectable items, see the AI Guardrails Console.	Use this service to detect whether AI-generated images contain noncompliant or unsuitable content. Apply it to all AI-generated images. For more information, see Image Moderation Enhanced 2.0 AIGC scenario detection service.
	AIGC Image Risk Detection (Outside Chinese Mainland) (aigcCheck_cb)	For AI-generated images intended for international audiences, detects pornography, vulgar content, sexually explicit content, politically sensitive content, terrorist content, prohibited content, disturbing content, and inappropriate content. For a full list of detectable items, see the AI Guardrails Console.	Use this service to detect whether AI-generated images contain noncompliant or unsuitable content. Apply it to all AI-generated images intended for international audiences. For more information, see Image Moderation Enhanced 2.0 AIGC scenario detection service.
	AIGC Violation Detection (aigcViolationDetection)	Detects trademarks, special logos, people, and other elements in images to assess potential infringement risks.	Use this service to assess infringement risks when applying AI-generated images—especially in commercial contexts. For more information, see Image Moderation Enhanced 2.0 AIGC scenario detection service.
	AIGC image detection (aigcDetector)	Input: Images. Detection: Whether images are likely AI-generated or synthetic.	Use this service to determine whether images are AI-generated. Apply it to tag image sources. For more information, see Image Moderation Enhanced 2.0 AIGC scenario detection service.
	AI-Generated Image Detection_Implicit Identifier (aigcDetectorFull)	Input: Images. Detection: Whether images are likely AI-generated or synthetic, and whether they contain implicit AI-generation markers (if present, returns AIGC metadata).	Use this service to determine whether images are AI-generated or synthetic. Apply it when you need implicit marker detection and synthetic trace analysis. For more information, see Image Moderation Enhanced 2.0 AIGC scenario detection service.
	AI Image Detection (Video Screenshots) (aigcDetectorForFrame)	Input: Video frames. Detection: Whether video frames are likely AI-generated or synthetic.	Use this service to determine whether video frames are AI-generated or synthetic. Apply it when analyzing synthetic traces in videos. For more information, see Image Moderation Enhanced 2.0 AIGC scenario detection service.
	AIGC Detection_Professional Edition (aigcDetector_pro)	Input: Images. Detection: Whether images are likely AI-generated or synthetic, and whether they show signs of synthesis (including face synthesis).	Use this service to determine whether images are forged. Apply it for image authentication. For more information, see Image Moderation Enhanced 2.0 AIGC scenario detection service.
	AI-Generated Image Detection_Ultimate (aigcDetector_ultra)	Input: Images. Detection: Whether images are likely AI-generated or synthetic, whether they show signs of synthesis (including face synthesis), and whether they contain AI-based local edits.	Use this service to determine whether images are forged. Apply it when you need to assess whether images contain AI-based local edits. For more information, see Image Moderation Enhanced 2.0 AIGC scenario detection service.
Business scenarios	Profile Photo Check (profilePhotoCheck)	Optimized for profile pictures. Detects pornography, sexually explicit content, politically sensitive content, terrorist content, prohibited content, religious content, advertising traffic diversion, special logos, behaviors, specific objects, inappropriate content, abuse, and image quality issues. Includes both visual content and embedded text. For a full list of detectable items, see the AI Guardrails Console.	Use this service to detect whether profile pictures contain noncompliant, unsuitable, or platform-disruptive content. Apply it to all profile picture images.
	Post Comment Image Moderation (postImageCheck)	Optimized for post and comment images. Detects pornography, sexually explicit content, politically sensitive content, terrorist content, prohibited content, religious content, advertising traffic diversion, special logos, behaviors, specific objects, and inappropriate content. Includes both visual content and embedded text. For a full list of detectable items, see the AI Guardrails Console.	Use this service to detect whether post and comment images contain noncompliant, unsuitable, or platform-disruptive content. Apply it to all post and comment images.
	Advertising Check (advertisingCheck)	Optimized for marketing materials. Detects advertising law violations, prohibited content, pornography, sexually explicit content, politically sensitive content, terrorist content, religious content, behaviors, inappropriate content, advertising traffic diversion, and specific objects. Includes both visual content and embedded text. For a full list of detectable items, see the AI Guardrails Console.	Use this service to detect whether marketing materials contain noncompliant, unsuitable, or platform-disruptive content. Apply it to all marketing and e-commerce images.
	Live Stream Check (liveStreamCheck)	Optimized for video and live stream screenshots. Detects pornography, sexually explicit content, politically sensitive content, terrorist content, prohibited content, religious content, inappropriate content, advertising traffic diversion, behaviors, specific objects, and image quality issues. Includes both visual content and embedded text. For a full list of detectable items, see the AI Guardrails Console.	Use this service to detect whether video and live stream screenshots contain noncompliant, unsuitable, or platform-disruptive content. Apply it to all video and live stream screenshots exposed to the public internet.
Special scenarios	Moderate Images for Malicious Content (riskDetection)	Detects whether images hide videos or players.	Use this service to detect malicious use of images to hide video segments or players. Prevent black-market theft of Object Storage Service and CDN traffic.

Note

Select the most appropriate service for your use case. For example, use Profile Picture Detection (profilePhotoCheck) for profile pictures. If your images do not fall into the profile picture, post and comment, AIGC, marketing material, or video screenshot categories, use Baseline Check (baselineCheck) to detect high-risk violations. If you also need to detect advertising traffic diversion, inappropriate guidance, or abuse, use Content Governance Detection (tonalityImprove) in addition.

2. Billing

Pay-as-you-go

Image Moderation Enhanced Edition uses pay-as-you-go billing by default. You are charged only when you call a service. The following table lists the API pricing.

Moderation type	Supported business scenarios (services)	Unit price
Standard image moderation (image_standard)	Baseline Check: baselineCheck Content Governance Detection: tonalityImprove AIGC Image Risk Detection: aigcCheck AIGC Image Infringement Detection: aigcViolationDetection AIGC Image Identification: aigcDetector	15 CNY per 10,000 calls Note One call to any service listed above counts as one billing unit. For example, 100 calls to Baseline Check cost 0.15 CNY.
Advanced image moderation (image_advanced)	Baseline Check (Professional Edition): baselineCheck_pro Baseline Check (International): baselineCheck_cb Content Governance Detection (International): tonalityImprove_cb AIGC Image Risk Detection (International): aigcCheck_cb AIGC Image Identification (with Implicit Markers): aigcDetectorFull AIGC Image Identification (Video Frame): aigcDetectorForFrame AIGC Image Identification (Professional Edition): aigcDetector_pro Profile Picture Detection: profilePhotoCheck Post and Comment Image Detection: postImageCheck Marketing Material Detection: advertisingCheck Video and Live Stream Screenshot Detection: liveStreamCheck	30 CNY per 10,000 calls Note One call to any service listed above counts as one billing unit. For example, 100 calls to Baseline Check (Professional Edition) cost 0.3 CNY.
Large-model image moderation (image_vl_standard)	Image Moderation for Large and Small Model Integration: postImageCheckByVL Image Moderation for Large and Small Model Integration_For regions outside the Chinese mainland: postImageCheckByVL_cb Image Moderation Service Based on LLMs: baselineCheckByVL LLMLarge Model-Powered Ad Traffic Detection: adCheckByVL Image Moderation for Large and Small Model Integration_China Version	45 CNY per 10,000 calls Note One call to any service listed above counts as one billing unit. For example, 100 calls to Large and Small Model Integration Image Moderation cost 0.45 CNY.
Basic large-model image moderation (image_vl_basic)	When you enable large-model moderation capability for small-model services in the console, this billing item applies. For configuration steps, see the Console operation guide.	15 CNY per 10,000 calls Note When you enable large-model capability for small-model services, this billing item applies. Charges apply based on actual usage.
Basic image authentication (image_identify_standard)	AI-Generated Image Detection_Ultimate: aigcDetector_ultra	200 CNY per 10,000 calls Note One call to any service listed above counts as one billing unit. For example, 100 calls to AIGC Image Identification (Ultimate Edition) cost 2 CNY.
General malicious image detection (risk_standard)	Malicious Image Detection: riskDetection	1 CNY per 10,000 calls Note This service uses a usage-based billing model, where each invocation is a billable event. For example, 100 invocations of the General Baseline Detection Professional Edition cost CNY 0.01.

Note

Content Moderation 2.0 pay-as-you-go usage is billed every 24 hours. In the billing detail report, the moderationType field corresponds to the moderation type field described earlier. You can view Billing Details.

Resource plan deduction

If your moderation volume is large or predictable, purchase a resource plan. Larger plans offer lower per-unit costs than pay-as-you-go pricing. For more information, see Purchase Content Moderation Enhanced Edition resource plans.

Content Moderation Enhanced Edition resource plans offset usage of Content Moderation Enhanced Edition and cannot be shared with the Content Moderation 1.0 data transfer plan. The specific offset factors are as follows:

Moderation type	Offset factor
Standard image moderation (image_standard)	2: Each successful API call deducts 2 units from the resource plan quota. Note For example, if your resource plan quota is 10 units and you make 1 successful API call, 2 units are deducted, leaving 8 units.
Advanced image moderation (image_advanced)	4: Each successful API call deducts 4 units from the resource plan quota. Note For example, if your resource plan quota is 10 units and you make 1 successful API call, 4 units are deducted, leaving 6 units.
Large-model image moderation (image_vl_standard)	6: Each successful API call deducts 6 units from the resource plan quota. Note For example, if your resource plan quota is 10 units and you make 1 successful API call, 6 units are deducted, leaving 4 units.
Basic large-model image moderation (image_vl_basic)	2: Each successful API call deducts 2 units from the resource plan quota. Note For example, if your resource plan quota is 10 units and you make 1 successful API call, 2 units are deducted, leaving 8 units.
Basic image authentication (image_identify_standard)	26.67: Each successful API call deducts 26.67 units from the resource plan quota. Note For example, if your resource plan quota is 50 units and you make 1 successful API call, 26.67 units are deducted, leaving 23.33 units.
General malicious image detection (risk_standard)	0.13: Each successful API call deducts 0.13 units from the resource plan quota. Note For example, if your resource plan quota is 10 units and you make 1 successful API call, 0.13 units are deducted, leaving 9.87 units.

After purchase, Image Moderation Enhanced Edition API usage is deducted from your resource plan quota first. When the quota runs out, pay-as-you-go billing applies automatically. Monitor your remaining quota and charges. You can set low-balance alerts in the Resource Plan System in the Alibaba Cloud User Center.

3. Usage

Integration

Image Moderation 2.0 supports SDK integration and native HTTPS integration:

For SDK integration, see Image Moderation Enhanced Edition SDK and integration guide.
For native API integration:
- For large-model image moderation, see LLM Image Moderation.
- For general image detection, see Image moderation enhanced API.
- For AIGC and image authentication scenarios, see AIGC and Forgery Detection Service of Image Moderation Enhanced Edition.

Console operations

Before your first API call, configure your image moderation policy in the AI Guardrails Console.
In the console, you can adjust the detection scope, configure per-business policies, manage custom image libraries and dictionaries, and view call results and usage.
After adjusting your moderation policy, run effect testing in the console to verify the results.

For step-by-step instructions, see the Console